October 16, 2009

The emergence and dispersal of haplogroup J-P58 (aka J1e)

The paper uses the evolutionary mutation rate, which, as I have argued elsewhere overestimates time to most recent ancestor (TMRCA) by about a factor of 3. The evolutionary mutation rate is appropriate for haplogroups subject to strong genetic drift that have not grown to large numbers, but it is completely inappropriate under conditions of strong population growth.

To make things concrete, according to the model of drift-induced variance reduction proposed by Zhivotovsky, Underhill, and Feldman (2006), in 10,000 years (or 400 generations), J-P58 should have grown to the grand number of 200 men, or at least five orders of magnitude lower than the actual present-day haplogroup size. To account for the observed J-P58 size of millions of men, strong growth over time is needed, and with either the Z.U.F. (2006) analysis or my own, strong growth results in an accumulation of variance at close to the germline mutation rate.

With that said, all ages in this paper should be divided by a factor of 3. This is not only theoretically sound, but harmonizes better with other lines of evidence.

The paper studies Y-STR variance in several Middle Eastern populations. The lack of samples from the Caucasus does not allow us to infer the levels of Y-STR variance in that region. Arabian J-P58 from Saudi Arabia, Qatar, and UAE are pooled, resulting in low mean Y-STR variance of 0.16. This low value stems primarily from Qatar and UAE as the Saudi Arabian J-P58 makes a very small contribution (4 examples) in the pooled sample.

Unfortunately the authors just missed the very recent paper on Arabian DNA by Abu-Amero et al., which shows that J-M267 variance is 0.27-0.29 in Yemen and Saudi Arabia, and much lower (0.16-0.19) in UAE and Qatar. This severely weakens the case for an expansion of J1 from the northern to the southern Levant, as it reveals that not only Oman and Yemen (mentioned in the paper), but also the geographically dominant Saudi Arabia is a region of high Y-STR diversity. Thus it is not the case that:
The timing and geographical distribution of J1e is representative of a demic expansion of agriculturalists and herder–hunters from thePre-Pottery Neolithic B to the late Neolithic era.24,26 The higher variances observed in Oman, Yemen and Ethiopia suggest either sampling variability and/or demographic complexity associated with multiple founders and multiple migrations.
But rather Oman, Yemen and Ethiopia are not atypical for the southern J1e range, which also includes Saudi Arabia as a region of high Y-STR variance. It is rather only the small gulf states of UAE and Qatar that have lower variance.

An interesting find, however, is the fact of high Y-STR variance (0.37, 0.43) in Alawites from Syria and Assyrians from Syria and Iraq. These populations have an impeccable Semitic historical record, and, in the case of the Assyrians are one of the few non-Arabic populations included in the study. It is also interesting that Assyrians are said to be derived from both Assyrian- and Aramaic-speaking ancestors, and hence to potentially have a complex (both East- and Northwest- Semitic) origin. These facts probably explain their high Y-STR variance.

Translated into non-"evolutionary" years, the expansion time of 16.2ky for Assyrians, becomes ~5.4ky. This age is in uncanny agreement with the recently estimate age of Semitic languages 5.75ky ago.

The authors of the current paper cite the above-mentioned linguistic work, but have trouble bridging the gap between their own "evolutionary" dates and the date for the breakup of Proto-Semitic:
A recent Bayesian analysis of Semitic languages supports an originin the Levant 5750 years ago and subsequent arrival in the Horn of Africa from Arabia 2800 years ago,11 thus providing an indirect support of our phylogenetic clock estimates. It is important to note that the glottochronological dates yield estimates for the break-up and expansion of the Proto-Semitic language. Proto-Semitic, itself, may have been spoken in a localized linguistic community for millennia before its bifurcation into the East and West Semitic branches.
If one rejects the "evolutionary" rate, there is no need to postulate that Proto-Semitic was spoken (but did not disperse) for millennia; indeed, a "static" Proto-Semitic/J-P58 community would be difficult to explain in view of the fact that mobile herding was their main economic activity. In my view, The J-P58 bearing Proto-Semites emerge in the 4th millennium BC out of a general J1 Middle Eastern background, just as their TMRCA suggests. They begin to expand at that time, and emerge in the historical record 1-2 thousand years later in both their Eastern (Akkadian) and, later, Western (Aramaic and Canaanite) forms.

The authors also cite their own work with respect to the correlation of J1 distribution with semi-arid environments in the Middle East and cite evidence to the effect that:
archeological studies have shown an early presence (ca. 6000–7000 BCE) of domesticated herding in the arid steppe desert regions
The presence of a large frequency of undifferentiated J*(xJ1, J2) chromosomes in Soqotra suggests that the Arabian peninsula possessed such chromosomes, which now have a marginal status throughout the Middle East. I propose that a the early steppe desert herders of 6000-7000BC possessed J* chromosomes, that J1 arose in the Middle East, and its subclade J-P58 experienced rapid growth associated with the breakup and expansion of Semitic languages in the 4th millennium BC.

In conclusion: this paper gives us important new data on the origin and expansion of Y-chromosome J-P58, and strengthens the case that this haplogroup may be a diagnostic marker of the Proto-Semitic population of the Near East.


European Journal of Human Genetics doi: 10.1038/ejhg.2009.166

The emergence of Y-chromosome haplogroup J1e among Arabic-speaking populations

Jacques Chiaroni et al.


Haplogroup J1 is a prevalent Y-chromosome lineage within the Near East. We report the frequency and YSTR diversity data for its major sub-clade (J1e). The overall expansion time estimated from 453 chromosomes is 10 000 years. Moreover, the previously described J1 (DYS388=13) chromosomes, frequently found in the Caucasus and eastern Anatolian populations, were ancestral to J1e and displayed an expansion time of 9000 years. For J1e, the Zagros/Taurus mountain region displays the highest haplotype diversity, although the J1e frequency increases toward the peripheral Arabian Peninsula. The southerly pattern of decreasing expansion time estimates is consistent with the serial drift and founder effect processes. The first such migration is predicted to have occurred at the onset of the Neolithic, and accordingly J1e parallels the establishment of rain-fed agriculture and semi-nomadic herders throughout the Fertile Crescent. Subsequently, J1e lineages might have been involved in episodes of the expansion of pastoralists into arid habitats coinciding with the spread of Arabic and other Semitic-speaking populations.



Anonymous said...

The paper is on J1e in Arabic speaking populations of the Middle East, hence it is limited in its applications to other populations that have haplogroup J1e, which is the haplogroup, and subclade in the majority of J1 whether European Jews, Italian, Iberian, North African, Ethiopian and a sizable number of Anatolians, and Caucasus dwellers. According to Tofanelli et al, and as commented on by Cruciani et al many years ago (in genetic study terms) J1 is more diverse in Europeans, Anatolians and Ethiopians (Caucasus dwellers' haplogroups were not adequately tested at the time) than in any people's of the Middle East (outside of Turkey) except as Tofanelli said, Iran, "Kurdistan" and Lebanese Arabic speakers. Tofanelli even found, God forbid, that Ashkenazim J1 from Israel is more diverse than J1 of Iraqi, Algerians, many Caucasus dwellers and Sudanese. The study focused on Arabic speakers of the Middle East but used data from Anatolians, Kurds, Irani and Ethiopians to ascertain genetic diversity of J1e. The age of J1* has been estimated to be over 20 kya, similar to that of its cousin haplogroup I. The Arabic speakers are minor players in J1e, they just have a lot of closely related, numbers i.e frequency of the haplogroup and subclade. Frequency means nothing much other than high fertility, founder influences and cultural practices like marriage preferences. Arabians are notorious for their overemphasis on male lineages, prefering paternal cousin marriages and polygamy. Polygamy is sanctioned in the Quran.

Languages are loosely correlated with genetics, more of a coincidence than anything else. Languages are young compared with haplogroups,and learned. The irony is that languages are learned usually from one's mother not one's father. If a genetic correlation exists it should very evident with mtDNA haplogroups not Y haplogroups. A wise child knows its own father.

It is strange that Dienekes has complained about the overestimation of the age of J1 and other essentially rare European haplogroups (one's common in non European Caucasoid) with "too old, divide age by three" but says the opposite when R1 is found to be 18,500 years old and Siberian Asian in origin making it impossible for R1b to have a European foothold before the LGM. Eurocentrism. The most desired of Europeans, the tow heads of Scandinavia, are high haplogroup I, the first degree cousin of J1. What I know is this: I, J and G are western Eurasian. R is eastern Eurasian more in common with haplogroups found in Asia proper than Europe or the Middle East. It is most probable that Europeans are descended from people who entered Europe from the east in the last 20,000 years with the largest group coming in in the Holocene/Neolithic time.

Anonymous said...
This comment has been removed by a blog administrator.
ashraf said...

J1 is estimated 10-15 k years and
J1e is surely much more younger.

How it could be that R1 goes as long as R1b1ba2c3.. whereas J1's are only as short as J1a or J1b or J1e etc...

The European J1's could be Jews,in fact used to be many jews in Europe(Marx,Einstein,Freud etc..)and perhaps some phoenicians(as there was many Phoenician cities built on Europe mediterranean shores)

Maju said...

There are two reasons, Ashraf:

1. Unequal research of haplogroups. R1b is, in spite of the many blanks, the best researched haplogroup on Earth. However J and other haplogroups, most of them West Eurasian, are not that far behind.

2. That each haplogroup has a unique phylogeny and that of R1b has many small sub-branches along most of the rather thin but long stem, before the huge expansion at the R1b1b2a1(a) level in Europe. Instead J has an early bifucation (J1 and J2) followed by another one (J2a and J2b), whith most of the structure under these branches being scattered in a relatively homogeneous area, much like R1b1b2a1 is.

It's like comparing a climbing vine (R1b) with an orchard (J or J1) one. They are just different and they branch out mainly at very different heights from the ground, depending largely on where they have been cropped by the farmer (drift, founder effects).

Katharós said...

I am interested in what genetic context the Neolithic graves of Jebel Buhais “United Arab Emirates 5th millennium BC.” may stand to resent Y chromosome papers on the Arabian Peninsula. I don’t know much about BHS 18 my self, only that this burial site as about 400 well preserved individuals that were homogenous as a group, dating in the 5th millennium BC. and that they were nomadic pastoralists. If I understand the situation correctly in 4000BC. the Arabian Peninsula slowly starts to dry up and these people start to vanish from the site. Which is also documented by violence among these herders due to competition of habitat.

Especially this comment by Professor Hans-Peter Uerpmann the head of the project interests my.
Looking at the are of Jebel Buhais itself, the occurrence of Bronze Age burials of the Wadi Suq period (BHS 1;3;8:12:64;66) in close proximity to BHS 18 might indicate a long standing, local tradition of burial. Nevertheless there appears to be continuity neither in funerary practices nor between the populations. Comparing the skeletal morphology of the Neolithic population (BHS 18) and the Jebel Buhais Bronze Age population, a stronger, more robust form was observed in Neolithic times versus a more gracile form in the Wadi Suq “The Wadi Suq period (2000-1600 BC)”

Jebel Buhais

Katharós said...

And of course if there is a coherence between these Neolithic pastoralists and there disappearance in the 4th millennium BC. and expansion of Semitic languages in the 4th millennium BC.

On a site that is now down Hans-Peter Uerpmann remarks further studies on the whereabouts of these pastoralists is needed and that Mesopotamia might be the place to look for them.

Anonymous said...

ashraf if you believe that J1 is 10 to 15 ky old, you are a moron.

The early studies were heavily biased, very Eurocentric. The proved what they wanted to prove: R1b is Paleolithic, was present in the Cro Magnons and the early people who painted caves in France and Spain. J1 is considered to be Woggy, not European, belonging to the black arsed wogs of Arabia. Its age was fixed to a point where it couldn't possibly be found in Europe before the LGM or even in the period when Europeans supposedly all rushed out of Iberia or whatever European Eden to repopulate the real Europe, Central and North Europe.

Later studies have shown R1 to be 18,500 years old and Asian in origin not much different from its other Asian brother Q and its cousins O and N. R1b couldn't have been in Europe before the LGM and unless those men had winged feet, couldn't have been present in Europe much before Europe started to become habitable again. Haplogroup J like I is much older. J1 is at least 25 ky old. That is older than the LGM. What you say is that the R groups are unstable genetically as they have mutated a lot in a short period of time. That subclade of the Asiatic R1b you mentioned is about 4 ky old. That is the trouble with R1b. None of the subclades are old and mostly match the time of the entry into Europe of the Asian group of languages known as Indo-European which lead to the extinction of many native European languages.

There are not many subclades of the J group because up until now most of the genetic work has been done on the Asiatic R1b and the rest on haplogroup I. The haplogroup E subclade found in Europe has been extensively studied, mainly to separate it from the Wogs in Africa and the Middle East. All the kerfuffle about V13! What a lot of fuss about nothing. Haplogroup E is African doesn't matter where its subclades originate. Haplogroup J2 has been studied well as it not uncommon in Europeans (And Jews and Arabs). Just efforts to make it Whiter less Woggy, shifting its origins points to Anatolia or the Balkans instead of the Fertile Crescent where it originated. J1 has been sieved by everyone racist ratbags, Jews with their stupid Cohens, and the Muslim Arabs with their fiction about the supposed descendants of Shem like Kahtan or Adnan or Perished Arabs. J1 is really a minority haplogroup in Jews, less than 20%. In Arabians the high frequency of J1 means founder influences, cultural practices, and inbreeding. Yemenis have the highest rate of J1 and the highest rate of homogeneity. What Dienekes said about their J1 being highly diverse in not true. Most of their J1 would form a distinct star pattern showing descent from a few interrelated men. The highest J1 diversity is in Kurds I.E speakers, South Portuguese, I.E speakers, Iranians, I.E speakers, Ethiopians, mostly Semitic speakers and Anatolians, Altaic speakers. Where are the Arabic speakers and the Jews? Way down the list except for the Lebanese.

You say that the European J1 could be from Jews, and you mentioned the Phoenicians. Listen to this: European J1 is more genetically diverse than in any Jews. Jewish J1 is more diverse than in any Middle Eastern Arabic speaking groups. Lebanese J1 is more genetically diverse than in any groups from the Arabian peninsula. You can't say that about your R1b. It gets more diverse the further east from its frequency centre you go. Armenians and some Central Asians would kill any Basque on genetic diversity of their R1b.

The Semitic group of languages were in Africa 10 kya. The oldest record of any Semitic words was in Sumerian records, and can be seen entering the records, so the time of ingress of Semitic words, mostly names, can be dated. Unfortunately it does not go back to 10 ky. So the idea that the Semitic language has anything to do with haplogroup J1 is just unfounded. Look at the spread of the R1 subclades. It matches well the spread of I.E languages in Eurasia.

Maju, sorry mate but your website sucks.

Maju said...

Later studies have shown...

How have they "shown"? They are mere educated guesses, nothing else.

Maju, sorry mate but your website sucks.

Well, I do what I think best. At least I'm not a rabid racist.

ashraf said...

Afrasian languages homeland is fertile-crescent as proved by sumerian and hurrian loanwords in african branches of afrasian and the vocaulary of asian animals and snow+shared afrasian-indoeuropean common roots,grammar and pre-ibrahamic religions.
There was an important phoenician and jew colnisation in Europe(and also egyptians,berbers and arabs).
All sites give 10-15 ky for J1 and surely J1e is more younger.
R1 is more diverse in mideast than in europe and probably its origin is fertile crescent(whereas R2 origin is India)
For further explanations you could read my inerventions here:


Maju said...

I don't think it makes much sense to date J1, just a few nodes downstream of F, only 10-12,000 years ago and also I am concerned that the North African J1 would seem pretty old, certainly pre-Semitic, though if it is of Capsian origin, it could still be within that rough timeline you suggest. But if North African J1 subset (that is shared by some West Asians but distinct from the "Asian only" subset) expanded from Egypt with Capsian that must mean that overall J1 must be older.

Personally I'm more inclined to see all IJ major branches (except I1, which seems quite more recent) as evolved soon after the colonization of West Eurasia some 50-45,000 years ago (they'd be in the order of at least 30,000 y.o. therefore). But just my opinion anyhow.

However I agree that most of J1 in Europe, excepting the "Neolithic area" of the SE maybe, would seem like "recent" arrivals. But could be Berber for what I know: after all they have much more J1 than Jews.

RJ said...

Let it be. It's pseudoscience.

Anonymous said...

Ronojoy, you are not a scientist. So how do you know what is pseudoscience from your arse? Stick to knitting and take another tablet.

The dating of haplogroups and haplotypes is not exact as we do not have sufficient remains, and dna of good quality to find the SNPs which tell us the haplogroups of Paleolithic or Neolithic humans. That is a more authentic of the haplogroups that existed in those times than any calculation and statistics using data from modern populations which have shown itself to differ greatly from those Paleolithic/Neolithic humans. It is bloody hard work getting ancient dna in good quality from remains which have been in contact with soil and other organic materials for many thousands of years. Contamination in situ and post removal is a serious problem.

It is unfortunate for J1 that it is found in such high numbers ins despised groups of people in the Middle East or supposedly from the ME like Jews and Arabs, or in the visibly admixed Africans like North African Berbers and Amharic speaking Ethiopians. It is found also in Southern Europeans known to have a long history of occupation by non European Africans like Moors and Arabic speakers from Africa. It is natural to think of J1 as foreign, as a sign of Jews or Arabs or Moors. However science is not about what is natural but provable facts. J1 whatever its age happens to be more genetically diverse away from the populations with the highest frequency. Ethiopians have more diverse J1 than South Arabians, they also have more YCA 19/22 and DYS388 less than 17. Arabians in J1 are monotonously the same coming from a small group of related men. Ethiopians cannot be the spawned population from South Arabian men. The Ethiopians got their Caucasoid looks and their Semitic languages earlier and from a different group of J1 carrying men. It is likely the Ethiopian control of South Arabia and the reverse control of Ethiopia did little to change the already existing rates and types of J1 present in both populations. You cannot make a silk purse out of a sow's ear. The Arabians as far as J1 goes, but you can say that about any haplogroup found in them, R1a1, J2, C, the E subclades, are totally derivative of Northern populations. As J1e is the main subclade of J1, J1* being next, and found in all populations where J1 is present, its age has to be ancient. It is good a study concentrating on Arabic speakers has been done. It won't please the Arabians who think they populated the world with J1, and have an extensive mythohistory of the origins of Arabian tribes, the Kahtan/Adnan/Perished Arabs nonsense, but that is how science works: on facts which can be repeated. It would be ironic if the despised Sulubba tribe of Arabia turned out to have more diverse J1 than the true blue Arab the Bedouin. Such is life.

Katharós said...

@Ponto, if I am not mistaken you are also a J1 carrier. In case you are, where are your ancestors from? Because there are a few examples of Arabian mercenaries running around in Britain and Europe under Roman rule.

pconroy said...


Yeah that's possible. Ponto's J1 could also have come from the 5,500 Alano-Sarmatians settled just south of Hadrian's Wall. The Dagestanis have high J1 and they are situated in the North Caucasus right next to the Ossetians - descendants of the Alans. The Alano-Sarmatians have left traces of Steppe and Caucasus DNA in Lowland Scotland and Northern England - check out the Border Reiver website for more info:

Anonymous said...

@ pconroy : Aren't the Sarmatians and Alans (i.e. Scythians) supposed to be mainly of Y-DNA R1a1 ?

pconroy said...


Scythians were perhaps R1a1.

Alans and Sarmatians of the North Caucasus, live in an area that even today has considerable R1b

Maju said...

Aren't the Sarmatians and Alans (i.e. Scythians) supposed to be mainly of Y-DNA R1a1

In principle yes: they originated from a part of Europe where that lineage is clearly dominant. However the genetics of the Caucasus region is quite more complex and variegated and Ossetians (an Iranian language nation) are claimed to descend from Sarmathians and Alans.

Katharós said...

@ pconroy
Thanks for the information, I never heard about Sarmatian mercenaries in Britain. I was thinking of Syrian Bow men.
And other Middle Eastern input, see Arbeia a Roman fort in South Shields, Tyne & Wear, England.

Anonymous said...

pconroy : "Alans and Sarmatians of the North Caucasus, live in an area that even today has considerable R1b"

Indeed, but the Alans and Sarmatians were not living IN the Caucasus, they wrere living in Ukraine (and formerly Kazakhstan in ancient time dislodging later the "scythians", IIRC) then in east Europe (even north Balkans IIRC during the late Roman time). A mixing with peoples with haplogroups typical of the Caucasus is quite plausible but in large quantity, it's not sure at all.

Maju : "However the genetics of the Caucasus region is quite more complex and variegated and Ossetians (an Iranian language nation) are claimed to descend from Sarmathians and Alans."

Yes, but we can assume these formerly-"Alans" Ossetians are the remnants of a few Alans (the rest were part of the Germanic invasions in the Roman empire) that fled the Huns and Turkish waves in the mountains, but they could have been a influent minority imposing itself on the local populations.
I'm not sure the ossetians are representative of the Alans of yore (and even less Sarmatians).

Maju said...

I'm not sure the ossetians are representative of the Alans of yore (and even less Sarmatians).

Me neither - to be clear.

pconroy said...

Alans were living in the area around today's Hungary, and as far East as the Caucasus and beyond, having accompanied the Huns earlier on their invasion of Europe - being first conquered, then nominally in confederation with them.

In the 5th century after the Hun-Goth-Alan forces were beaten at the battle of Chalons in France by a Roman-Goth force, some Alans settled in Brittany. Later Alans accompanied the Vandals to Iberia and left their mark there too. The brought a large hunting dog with them, whose descendants in Spain today are called "Alano".

King Arthur and the Knights of the Round Table, are Alan folklore and that's why they exist in Brittany and also in Britain.

Maju said...

PConroy: the link between Britain and Brittany is that British (Celts from Cornwall and Devon) settled in Brittany at the early Middle Ages. That does not seem to be related at all with Alans. Making Arthur's legend be "Sarmathian" sounds very much far fetched to me, sincerely.

Katharós said...

Halfway off topic, still interesting.
"The Invention of the Jewish People"

Youtube vid. with Shlomo Sand
talking about his book. "The Invention of the Jewish People"

Anonymous said...

what about "The presence of a large frequency of undifferentiated J*(xJ1, J2) chromosomes in Soqotra suggests that the Arabian peninsula possessed such chromosomes" obviously j1 came from the south to south arabian peninsula and then norh ward to the bogus str388=13 of one man and his relatives. STR388 moves 3 steps at one time what is the big deal about 13 it is only one step from 16 of CMH. we the Arabs are not Gog and magog of the Caucasus, R1a and R1b the europpeans are!!

Maju said...

J(xJ1,J2) in a single location like Socotra is probably just another undescribed J sublineage (i.e. J3). While it adds weight to a West Asian origin of J and is very interesting in itself, it says absolutely nothing re. J1, only about J as a whole.

Most people would agree that the diversity of J1 in Arabia peninsula is low and that this one is higher further north, in Palestine for instance, which has the honor of being one of the homelands of Neolithic in West Eurasia and the World. North African J1 is also highly diverse and there is a STR subcluster in J1 (about half of the J1 population) that looks like coalescing in North Africa.

All that is best explained within the Epipaleolithic and Neolithic demographic flows, though I understand that J1 as such is older than that.

Arabia peninsula has also been demonstrated to have Fertile Crescent-derived lineages in mtDNA (per Abu Amero 2008). All this is logical if you consider the aridity of the region and the fact that it could barely be inhabited at all before the camel was domesticated, which is pretty late.

Gog and Magog are silly mythological fantasies that have no place in any scientific discussion.